[ROCm] Enable Aiter ck_gemm_a8w8_blockscale for RDNA4 gpus. Qwen3.5-27B-FP8 tp=2, Qwen3-0.6B-FP8 tp=1 #77
Open
big-yellow-duck wants to merge 10 commits intomainfrom
Open
[ROCm] Enable Aiter ck_gemm_a8w8_blockscale for RDNA4 gpus. Qwen3.5-27B-FP8 tp=2, Qwen3-0.6B-FP8 tp=1 #77big-yellow-duck wants to merge 10 commits intomainfrom
big-yellow-duck wants to merge 10 commits intomainfrom